Power distribution area terminal configuration and dynamic deduction method, device, equipment and medium

Abstract

The invention provides a power distribution area terminal configuration and dynamic deduction method, which comprises the following steps: acquiring a first topological structure of a low-voltage power distribution area, an installation position of measurement equipment, electrical information acquired by the measurement equipment and a terminal configuration node, and arranging a feeder terminal device in the low-voltage power distribution area; performing initialization operation on the low-voltage monitoring unit, assigning a virtual load of the low-voltage power distribution area to obtain a measurement map, performing observability analysis, determining an observable range of the low-voltage power distribution area, obtaining a configuration scheme of the feeder terminal device, obtaining first measurement data of the low-voltage power distribution area, performing time mark alignment on the first measurement data, and obtaining a second measurement data of the low-voltage power distribution area; and obtaining second measurement data, establishing a measurement database of the low-voltage distribution area, and establishing a state deduction model according to the measurement database. According to the technical scheme of the embodiment, the comprehensiveness of measurement data acquisition of the low-voltage transformer area can be improved, and real-time dynamic deduction of the state of the transformer area is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power resource optimization, and in particular to a method, device, equipment and medium for terminal configuration and dynamic deduction in a distribution station area. Background Art

[0002] The measuring equipment in the low-voltage distribution substation area in the prior art may fail or be interfered with in the data collection, measurement and conversion links. In addition, due to the presence of various measuring equipment in the low-voltage distribution substation area, there are large differences in the sampling time scale, measurement accuracy and other aspects, resulting in the measurement information of the low-voltage distribution substation area being out of sync and low data utilization. In the case of insufficient real-time measurement, the evaluation accuracy is poor and cannot meet the expected timeliness requirements. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method, device, equipment and medium for terminal configuration and dynamic deduction of distribution substations, which can improve the comprehensiveness of low-voltage substation measurement data collection and realize real-time state deduction of substation status.

[0004] In a first aspect, an embodiment of the present invention provides a method for configuring and dynamically deducing a terminal in a distribution station area, including:

[0005] Acquire a first topological structure of a low-voltage distribution station area, an installation location of a measuring device, electrical information collected by the measuring device, and a terminal configuration node, wherein the low-voltage distribution station area is provided with a feeder terminal device;

[0006] According to the first topology structure, the installation location and the electrical information supplement zero injection node, the low voltage monitoring unit is initialized, and the virtual load of the low voltage distribution station area is assigned according to the terminal configuration node to obtain a measurement diagram of the low voltage distribution station area;

[0007] Constructing an observable tree search method for the low-voltage distribution station area to perform observability analysis on the measurement diagram to determine the observable range of the low-voltage distribution station area, and obtaining a configuration scheme of the feeder terminal device according to the observable range;

[0008] Acquire first measurement data of the low-voltage distribution station area according to the configuration scheme, perform time-scale alignment on the first measurement data, and obtain second measurement data;

[0009] A measurement database of the low-voltage distribution station area is established according to the second measurement data, and a state deduction model is established according to the measurement database to complete a dynamic real-time evaluation of the low-voltage distribution station area.

[0010] In some embodiments of the present invention, the observable tree search method for constructing the low voltage distribution station area performs observability analysis on the measurement map, including:

[0011] Confirm multiple equivalent connection nodes in the measurement graph, assign edge weights to the equivalent connection nodes, obtain a first node set, a first edge weight value, and the number of nodes of the measurement graph, and when the first node set and the first edge weight value are equal to the number of nodes, obtain the maximum observable range of the measurement graph;

[0012] When the node set and the edge weight value are not equal to the number of nodes, confirm multiple first isolated points in the measurement diagram, search for a first boundary connected to the first isolated point, and when there is reactive power flow measurement at the first boundary, obtain the second edge weight value and the second node set until all the first isolated points are searched, and when the second variable weight value and the second node set are equal to the number of nodes, obtain the maximum observable range of the measurement diagram;

[0013] Confirm a plurality of second isolated points in the measurement diagram, and when the reactive power flow measurement does not exist between the second isolated point and the first boundary, obtain a first reactive injection measurement node and a third node set from the nodes, and according to the first reactive injection node and the third node set;

[0014] Obtain a fourth node set of the node, when there is no reactive power flow measurement at all the boundaries, obtain a second reactive injection node, when there is a second boundary connected to the second reactive injection node, and the second boundary and the second reactive injection node have the reactive power flow measurement, obtain a fourth node set, when the second boundary and the second reactive injection node do not have the reactive power flow measurement, continue searching until all the nodes are searched;

[0015] When the first node set, the second node set, the third node set and the fourth node set are all equal to the number of nodes, the maximum observable range of the measurement graph is obtained;

[0016] When the low-voltage distribution station is divided into a plurality of observable areas, the plurality of observable areas output their respective maximum observable ranges.

[0017] In some embodiments of the present invention, obtaining a configuration scheme of the feeder terminal device according to the observable range includes:

[0018] Acquire a virtual power flow value of the feeder terminal device, and obtain an objective function of the feeder terminal device according to the virtual power flow value;

[0019] Acquire the virtual power flow of the low-voltage distribution substation area, and construct the virtual power flow constraint of the low-voltage distribution substation area according to the objective function and the virtual power flow;

[0020] Acquire the branch set of the low-voltage distribution station area and the configuration status information of the feeder terminal device, and construct a configuration quantity constraint of the feeder terminal device according to the configuration status information of the branch set;

[0021] Acquire the virtual load of the node of the low-voltage distribution substation, calculate the line virtual power flow of the low-voltage distribution substation according to the virtual load and the virtual power flow, and construct the connectivity constraint of the low-voltage distribution substation according to the virtual load, the virtual power flow and the line virtual power flow;

[0022] Acquire virtual branch variable information of the low-voltage distribution station area and quantity information of the feeder terminals, and construct radial constraints of the low-voltage distribution station area according to the virtual branch variable information and the quantity information;

[0023] The upper limit constraint of the branch virtual power flow in the low-voltage distribution station area is constructed according to the quantity information, the configuration quantity constraint and the connectivity constraint, and the lower limit constraint of the branch virtual power flow in the low-voltage distribution station area is constructed according to the virtual power flow constraint, the quantity information and the radial constraint.

[0024] In some embodiments of the present invention, after obtaining the first measurement data of the low-voltage distribution station area according to the configuration scheme, the method further includes:

[0025] Acquire the first measurement data and preset basic data information of all the measurement devices, and perform a normalization operation on the first measurement data according to the basic data information to obtain a normalization result of the first measurement data;

[0026] Acquire a characteristic curve of the first measurement data, wherein the characteristic curve determines an optimal solution of cluster centers and the number of clusters in the normalization result according to a preset first function and a first algorithm, and identify the first measurement data according to the optimal solution;

[0027] Acquire the data category, sample number, clustering parameter and fuzzy matrix of the first measurement data, and construct an indicator function of the first measurement data according to the data category, sample number, clustering parameter, cluster center, cluster number and fuzzy matrix;

[0028] Obtaining a probability partition matrix of the first measurement data and a normalized sample number of the normalized result, constructing a membership matrix according to the probability partition matrix and the normalized sample number, and confirming the relevance of the first measurement data according to confirming the data category;

[0029] The membership matrix replaces the fuzzy matrix, and introduces the probability partition matrix into the first function, wherein the fuzzy matrix meets the normalization condition, so as to obtain the normalization constraint of the first measurement data;

[0030] A clustering result of the first measurement data is obtained according to the normalization constraint and the indicator function.

[0031] In some embodiments of the present invention, establishing a measurement database of the low-voltage distribution station area according to the second measurement data includes:

[0032] Obtaining a ring adjacency matrix, a degree matrix, a first output value, and an activation function of the low-voltage distribution station area, and constructing a GCN model according to the ring adjacency matrix, the degree matrix, the first output value, and the activation function;

[0033] Obtaining the reset gate, update gate, hidden layer and training weight of the GCN model at the first moment, and using the reset gate, update gate, hidden layer and training weight of the GRU model to process the timing information of the low-voltage distribution station area;

[0034] confirming whether there is bad data in the second measurement data according to a second algorithm;

[0035] When the bad data does not exist in the second measurement data, directly establishing a measurement database;

[0036] When the bad data exists in the second measurement data, data reconstruction is performed on the second measurement data according to the GCN model and the GRU model.

[0037] In some embodiments of the present invention, the step of performing time-scale alignment on the first measurement data includes:

[0038] Obtaining the first measurement sequence number, the first sampling period and the first measurement value of the branch monitoring unit of the low-voltage distribution station area after the second moment of the low-voltage distribution station area;

[0039] Performing measurement data fusion on the branch monitoring unit according to the first measurement sequence number, the first sampling period and the first measurement value;

[0040] Obtaining the second measurement sequence number, the second sampling period and the second measurement value of the photovoltaic switch collection device in the low-voltage distribution station area after the second moment in the low-voltage distribution station area;

[0041] Performing measurement data fusion on the photovoltaic switch acquisition device according to the second measurement sequence number, the second sampling period and the second measurement value;

[0042] Obtaining a third measurement sequence number, a third sampling period, and a third measurement value of the multi-function meter in the low-voltage distribution station area after the second moment in the low-voltage distribution station area;

[0043] Performing measurement data fusion on the multi-function meter according to the third measurement sequence number, the third sampling period and the third measurement value;

[0044] Obtaining a fourth measurement sequence number, a fourth sampling period, and a fourth measurement value of the smart electric meter in the low-voltage distribution station area after the second moment in the low-voltage distribution station area;

[0045] The measurement data of the smart meter is fused according to the fourth measurement sequence number, the fourth sampling period and the fourth measurement value.

[0046] In some embodiments of the present invention, establishing a state deduction model according to the measurement database includes:

[0047] Obtaining a horizontal component, a vertical component, a first smoothing parameter, a second smoothing parameter, and a first predicted value and a first estimated value of a state quantity at a third moment of the low-voltage distribution station area;

[0048] Constructing a state equation of the low voltage distribution station area according to the horizontal component, the vertical component, the first smoothing parameter, the second smoothing parameter, the first predicted value and the second predicted value;

[0049] Obtaining the phase angle difference, active injection power, reactive injection power, mutual admittance between the multiple nodes, active power at the beginning of the line, and reactive power at the beginning of the line of multiple nodes in the low-voltage distribution station area;

[0050] Constructing a measurement equation of the low-voltage distribution station area according to the phase angle difference, the active injection power, the reactive injection power, the mutual admittance, the active power at the beginning of the line and the reactive power at the beginning of the line;

[0051] Obtaining the sampling point mean, sampling point weight, proportional correction factor, free parameter, influencing factor, state quantity mean and first covariance of the low voltage distribution station area;

[0052] Selecting a Sigma point according to the sampling point mean, the sampling point weight, the proportional correction factor, the free parameter, the influencing factor, the state quantity mean and the first covariance, and constructing a system prediction step model according to the Sigma point;

[0053] According to the system prediction unit model, the state quantity mean, state quantity covariance, Sigma point measurement value, third prediction value, measurement prediction covariance and second covariance are obtained, wherein the second covariance is the difference between the first prior state estimation value of the low-voltage distribution station area and the measurement prediction covariance;

[0054] Constructing a measurement update step model according to the state quantity mean, the state quantity covariance, the Sigma point measurement value, the third predicted value, the measurement prediction covariance and the second covariance;

[0055] Obtaining a second a priori prediction value, a third a priori prediction value, an actual measurement value and a posteriori estimation value of the low voltage distribution station area;

[0056] A state deduction model is constructed according to the state equation, the measurement equation, the prediction step model, the update unit model, the second priori prediction value, the third priori prediction value, the actual measurement value and the posterior estimation value.

[0057] In a second aspect, an embodiment of the present invention provides a distribution station area data status deduction device, comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the distribution station area terminal configuration and dynamic deduction method as described in the first aspect above.

[0058] In a third aspect, an embodiment of the present invention provides an electronic device, comprising the distribution station area data status deduction device as described in the second aspect above.

[0059] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the distribution station area terminal configuration and dynamic deduction method as described in the first aspect above.

[0060] The method for configuring and dynamically deducing terminals in a distribution area according to an embodiment of the present invention has at least the following beneficial effects:

[0061] Obtain a first topological structure of a low-voltage distribution station area, an installation position of a measuring device, electrical information collected by the measuring device, and a terminal configuration node, wherein the low-voltage distribution station area is provided with a feeder terminal device; supplement zero injection nodes according to the first topological structure, installation position, and electrical information, initialize the low-voltage monitoring unit, and assign a virtual load to the low-voltage distribution station area according to the terminal configuration node to obtain a measurement diagram of the low-voltage distribution station area; construct an observable tree search method for the low-voltage distribution station area to perform observability analysis on the measurement diagram to determine the observable range of the low-voltage distribution station area, and obtain a configuration scheme of the feeder terminal device according to the observable range; obtain first measurement data of the low-voltage distribution station area according to the configuration scheme, time-align the first measurement data, and obtain second measurement data; establish a measurement database of the low-voltage distribution station area according to the second measurement data, establish a state deduction model according to the measurement database, and complete a dynamic real-time evaluation of the low-voltage distribution station area. According to the technical solution of this embodiment, it is possible to effectively reduce the number of feeder terminal devices configured on the basis of ensuring that the substation is completely observable, realize the minimized collection of low-voltage distribution substations, so that the data deduction of low-voltage distribution substations can provide a more reliable data source and improve the comprehensiveness of substation measurement data collection. In this embodiment, the PSO algorithm is used to optimize the PFCM clustering algorithm to form a PSO-PFCM algorithm. In the process of measuring data recognition, the number of clusters and cluster centers can be optimized. The bad data in the measurement data can be intelligently identified through the characteristic curve of the measurement data. There is no need to construct the error covariance matrix of the measurement equipment. Only the measurement data amount is counted and analyzed, which significantly reduces the complexity of bad data recognition. In addition, this embodiment adopts the T-GCN measurement data reconstruction method. Compared with the ordinary time series neural network that only considers the time series correlation of the measurement data, the T-GCN neural network used in the present invention not only considers the time series of the measurement data, but also considers the spatial relationship between nodes, and further considers the coupling correlation between the measurement data. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a flow chart of data status deduction of a distribution station area provided by an embodiment of the present invention;

[0063] Figure 2 It is a flow chart of the observable tree search method for constructing a low-voltage distribution station area provided by the present invention to perform observability analysis on the measurement diagram;

[0064] Figure 3 It is a flow chart of a configuration scheme of a feeder terminal device obtained according to an observable range provided by the present invention;

[0065] Figure 4 It is a flow chart after obtaining the first measurement data of the low-voltage distribution station area according to the configuration scheme provided by the present invention;

[0066] Figure 5 is a flow chart of establishing a measurement database of a low-voltage distribution station area according to second measurement data provided by the present invention;

[0067] Figure 6 is a flow chart of performing time-scale alignment on first measurement data provided by the present invention;

[0068] Figure 7 It is a flow chart of establishing a state deduction model according to a measurement database provided by the present invention;

[0069] Figure 8 is a structural diagram of a data state deduction device for a distribution station area provided by another embodiment of the present invention;

[0070] Fig. 9 A schematic diagram of the structure of the T-GCN model provided by the present invention;

[0071] Fig.10 A schematic diagram of the GCN principle provided by the present invention;

[0072] Fig.11 This is a schematic diagram of the T-GCN unit structure provided by the present invention. DETAILED DESCRIPTION

[0073] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0074] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0075] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0076] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0077] An embodiment of the present invention provides a method for terminal configuration and dynamic deduction of a distribution station area, including obtaining a first topological structure of a low-voltage distribution station area, an installation position of a measuring device, electrical information collected by the measuring device, and a terminal configuration node, wherein the low-voltage distribution station area is provided with a feeder terminal device; a zero injection node is supplemented according to the first topological structure, the installation position, and the electrical information, an initialization operation is performed on a low-voltage monitoring unit, and a virtual load of the low-voltage distribution station area is assigned according to the terminal configuration node to obtain a measurement diagram of the low-voltage distribution station area; an observable tree search method of the low-voltage distribution station area is constructed to perform observability analysis on the measurement diagram to determine the observable range of the low-voltage distribution station area, and a configuration scheme of the feeder terminal device is obtained according to the observable range; first measurement data of the low-voltage distribution station area is obtained according to the configuration scheme, and the first measurement data is time-aligned to obtain second measurement data; a measurement database of the low-voltage distribution station area is established according to the second measurement data, and a state deduction model is established according to the measurement database to complete dynamic real-time evaluation of the low-voltage distribution station area. According to the technical solution of this embodiment, it is possible to effectively reduce the number of feeder terminal devices configured on the basis of ensuring that the substation is completely observable, realize the minimized collection of low-voltage distribution substations, so that the data deduction of low-voltage distribution substations can provide a more reliable data source and improve the comprehensiveness of substation measurement data collection. In this embodiment, the PSO algorithm is used to optimize the PFCM clustering algorithm to form a PSO-PFCM algorithm. In the process of measuring data recognition, the number of clusters and cluster centers can be optimized, and the bad data in the measurement data can be intelligently identified through the characteristic curve of the measurement data. There is no need to construct the error covariance matrix of the measurement equipment. Only the measurement data amount is counted and analyzed, which significantly reduces the complexity of bad data recognition. In addition, this embodiment adopts the T-GCN measurement data reconstruction method. Compared with the ordinary time series neural network that only considers the time series correlation of the measurement data, the T-GCN neural network used in the present invention not only considers the time series of the measurement data, but also considers the spatial relationship between nodes, and further considers the coupling correlation between the measurement data.

[0078] The control method of the embodiment of the present invention is further described below based on the accompanying drawings.

[0079] Reference Figure 1 , Figure 1 A flowchart of a method for configuring and dynamically deducing a terminal in a distribution area provided by an embodiment of the present invention includes but is not limited to the following steps:

[0080] Step S11, obtaining a first topological structure of a low-voltage distribution station area, an installation location of a measuring device, electrical information collected by the measuring device, and a terminal configuration node, wherein the low-voltage distribution station area is provided with a feeder terminal device;

[0081] It should be noted that the topological structure of the low-voltage distribution substation includes the substation layer, feeder layer and user layer. The substation layer is centered on the distribution transformer and radiates around with the power supply distance as the radius to form a power supply range or area. The feeder layer contains branch lines starting from the distribution transformer, and the branch lines are monitored by feeder terminal devices. The user layer is the final power consumption unit, which is monitored by terminal sensing or terminals.

[0082] Step S12, supplementing the zero injection node according to the first topological structure, the installation location and the electrical information, initializing the low voltage monitoring unit, and assigning the virtual load of the low voltage distribution station area according to the terminal configuration node to obtain a measurement diagram of the low voltage distribution station area;

[0083] It should be noted that in order to facilitate the distinction of branches, box transformers, branch boxes and incoming and outgoing line switches of distribution lines in the low-voltage distribution station area, the branch points of the box transformers, branch boxes and meter boxes are supplemented with zero injection nodes, the number of low-voltage monitoring unit configurations is initialized to 0, and the virtual load is assigned according to the acquisition terminal configuration node, and the connectivity diagram, reactive power measurement and voltage amplitude measurement of the low-voltage distribution station area are obtained to form a measurement diagram of the low-voltage distribution station area.

[0084] Step S13, constructing an observable tree search method for the low-voltage distribution station area to perform observability analysis on the measurement diagram to determine the observable range of the low-voltage distribution station area, and obtaining a configuration scheme of the feeder terminal device according to the observable range;

[0085] It should be noted that the observable tree search method can describe the network structure of the low-voltage distribution substation in detail, so as to accurately identify the connection relationship between each node and branch. By constructing the observable tree search method, all key nodes and branches in the low-voltage distribution substation can be fully covered, thereby ensuring the observable range of the distribution system and obtaining the configuration plan of the feeder terminal device based on the observable range.

[0086] Step S14, acquiring first measurement data of the low-voltage distribution station area according to the configuration scheme, and performing time-stamp alignment on the first measurement data to obtain second measurement data;

[0087] It should be noted that the first measurement data of the low-voltage distribution station area is obtained according to the configuration scheme, wherein the first measurement data includes node voltage, line current, line active power and reactive power flow, and then the first measurement data is normalized to obtain multiple measurement data characteristic curves. Based on the telemetry data and the recursive average filtering method, the first measurement data is time-aligned to obtain the second measurement data.

[0088] Step S15, establishing a measurement database of the low-voltage distribution substation according to the second measurement data, establishing a state deduction model according to the measurement database, and completing a dynamic real-time evaluation of the low-voltage distribution substation.

[0089] It should be noted that, by integrating the second-hand measurement data processed by various measuring devices, the dynamic state evaluation of the low-voltage distribution station area is carried out based on the unscented Kalman filter, and the dynamic real-time evaluation of the low-voltage distribution station area with multi-source measurement data preprocessing is established.

[0090] It should be noted that this embodiment can effectively reduce the number of feeder terminal devices configured on the basis of ensuring that the low-voltage distribution station area is fully observable, and realize the minimization of the area collection. This provides a more reliable data source for low-voltage distribution station area data deduction and improves the comprehensiveness of the area measurement data collection. And on the basis of the optimized configuration of the area measurement equipment and data preprocessing, the effectiveness of the measurement equipment optimization configuration model proposed in this embodiment is proved in the form of data, and the real-time dynamic estimation of the area status is realized to obtain a higher precision estimation result.

[0091] In addition, in one embodiment, referring to Figure 2 ,exist Figure 1 Step S13 of the illustrated embodiment also includes but is not limited to the following steps:

[0092] Step S21, confirm multiple equivalent connection nodes in the measurement graph, assign edge weights to the equivalent connection nodes, obtain a first node set, a first edge weight value, and the number of nodes of the measurement graph, and when the first node set and the first edge weight value are equal to the number of nodes, obtain the maximum observable range of the measurement graph;

[0093] Step S22, when the node set and the edge weight value are not equal to the number of nodes, confirm multiple first isolated points in the measurement diagram, search for the first boundary connected to the first isolated point, and when there is reactive power flow measurement on the first boundary, obtain the second edge weight value and the second node set, until all the first isolated points are searched, and when the second variable weight value and the second node set are equal to the number of nodes, obtain the maximum observable range of the measurement diagram;

[0094] Step S23, confirming a plurality of second isolated points in the measurement diagram, when there is no reactive power flow measurement between the second isolated point and the first boundary, obtaining a first reactive injection measurement node and a third node set in the node, according to the first reactive injection node and the third node set;

[0095] Step S24, obtaining a fourth node set of nodes, when there is no reactive power flow measurement at all boundaries, obtaining a second reactive injection node, when there is a second boundary connected to the second reactive injection node, and there is reactive power flow measurement between the second boundary and the second reactive injection node, obtaining a fourth node set, when there is no reactive power flow measurement between the second boundary and the second reactive injection node, continuing to search until all nodes are searched;

[0096] Step S25, when the first node set, the second node set, the third node set and the fourth node set are all equal to the number of nodes, the maximum observable range of the measurement graph is obtained;

[0097] Step S26, when the low voltage distribution station area is divided into multiple observable areas, the multiple observable areas output their respective maximum observable ranges.

[0098] It should be noted that according to the connected graph G n , reactive power measurement, voltage amplitude measurement to form measurement diagram G m , assign the edge weight connected to the equivalent ground node to 1, that is, K n+1,j =1,j∈Ω n , if the first formula is satisfied, the maximum observable range is output, otherwise search downward.

[0099] The first formula is as follows:

[0100]

[0101] Where: Ω n is the set of all nodes; K i,j is the weight value of edge ij; n is the number of nodes.

[0102] Select the first isolated point u, search for the first boundary uj connected to the first isolated point, if there is reactive power flow measurement on the first boundary uj, then the first boundary The first node set is the first node that has reactive power flow measurement on the first boundary uj and is connected to the first isolated point u. Then turn to the remaining isolated points and repeat the above process until all isolated points are searched. If the second formula is satisfied, the maximum observable range is output, otherwise search downward.

[0103] The second formula is:

[0104]

[0105] Select a second isolated point u, where there is no flow measurement on the first boundary connected to it, that is, the first boundary K u,j = 0, then find the second node v where node reactive power injection measurement exists at node j, the second node is a second node set consisting of the second nodes connected to the second isolated point u and having reactive injection measurement. Then find the edge vj connected to the second node v and having reactive power flow measurement. is the third node set consisting of nodes with reactive power flow measurement on edge vj and connected to the second node v. If all edges vj do not have reactive power flow measurement, then find the first reactive power measurement node w in j. If there is an edge wj connected to w and it has reactive power flow measurement, then The fourth node set is composed of nodes that have reactive power flow measurement on edge wj and are connected to node w. If all edges wj do ​​not have reactive power flow measurement, the search continues according to the same rule until all nodes are searched. If the third formula is satisfied, the maximum observable range is output, otherwise the search continues downward.

[0106] The third formula is:

[0107]

[0108] If there are still s nodes z with reactive injection measurement, then the edge connected to this node Determine whether the fourth formula is satisfied.

[0109] The fourth formula is:

[0110]

[0111] Output whether the low-voltage distribution station area is observable and output the maximum observable range. When the first node set, the second node set, the third node set and the fourth node set are all equal to the number of nodes, if the station area is divided into several observable areas, output their respective observable ranges.

[0112] In addition, in one embodiment, referring to Figure 3 ,exist Figure 1 Step S13 of the illustrated embodiment also includes but is not limited to the following steps:

[0113] Step S31, obtaining a virtual power flow value of the feeder terminal device, and obtaining an objective function of the feeder terminal device according to the virtual power flow value;

[0114] Step S32, obtaining a virtual power flow of the low-voltage distribution substation area, and constructing a virtual power flow constraint of the low-voltage distribution substation area according to the objective function and the virtual power flow;

[0115] Step S33, obtaining the configuration status information of the branch set and the feeder terminal device in the low-voltage distribution station area, and constructing the configuration quantity constraint of the feeder terminal device according to the branch set configuration status information;

[0116] Step S34, obtaining the virtual load of the nodes in the low-voltage distribution substation area, calculating the line virtual power flow of the low-voltage distribution substation area according to the virtual load and the virtual power flow, and constructing the connectivity constraint of the low-voltage distribution substation area according to the virtual load, the virtual power flow and the line virtual power flow;

[0117] Step S35, obtaining virtual branch variable information and quantity information of feeder terminals in the low-voltage distribution station area, and constructing radial constraints of the low-voltage distribution station area according to the virtual branch variable information and quantity information;

[0118] Step S36, construct the upper limit constraint of the branch virtual power flow of the low-voltage distribution station area according to the quantity information, configuration quantity constraint and connectivity constraint, and construct the lower limit constraint of the branch virtual power flow of the low-voltage distribution station area according to the virtual power flow constraint, quantity information and radial constraint.

[0119] It should be noted that this embodiment adopts the objective function of the LTU optimization configuration model based on the single commodity flow concept, wherein the objective function can be expressed by the following fifth formula:

[0120] F=minf max ;

[0121] Where F is the objective function and fmax is the maximum value of all virtual power flows.

[0122] The partition maximum virtual power flow constraint is constructed based on the virtual power flow and the objective function, which is expressed by the following sixth formula:

[0123] f o,j ≤f max ,j∈Ω n ;

[0124] Where: f o,j is the virtual power flow transmitted from the virtual branch oj to the node j.

[0125] By obtaining the branch set and the configuration status information of the feeder terminal device in the low-voltage distribution station area, and constructing the configuration quantity constraint of the feeder terminal device according to the branch set and the configuration status information, it is expressed by the following seventh formula:

[0126]

[0127] Among them, Ωl is the branch set consisting of all branches; D ij is a 0-1 variable representing the configuration status of the feeder terminal device of line ij, D ij =1 indicates configuration, D ij =0 means not configured.

[0128] The virtual loads of multiple nodes in the low-voltage distribution substation are obtained, and the line virtual power flow of the low-voltage distribution substation is calculated according to the virtual load and virtual power flow. The connectivity constraint of the low-voltage distribution substation is constructed according to the virtual load, virtual power flow and line virtual power flow, which is expressed by the following eighth formula:

[0129]

[0130] in: is the virtual load of node j, the virtual load of the node without measurement is 1, and the virtual load of the node without measurement is 1 / n; f ij is the virtual power flow transmitted on line ij.

[0131] The virtual branch variable information and the quantity information of the feeder terminal devices in the low-voltage distribution substation are obtained, and the radial constraint of the low-voltage distribution substation is constructed according to the virtual branch variable information and the quantity information, which is expressed by the following ninth formula:

[0132]

[0133] Where: l o,j is a 0-1 variable indicating whether there is a virtual branch. o,j =1 means that node j is connected to the virtual power source through a virtual branch, l o,j =0 means that node j is not connected to the virtual power supply.

[0134] According to the quantity information, configuration quantity constraints and connectivity constraints, the upper limit constraint of the branch virtual power flow in the low-voltage distribution station area is constructed, which is expressed by the following tenth formula:

[0135] -(nm)(1-D ij )≤f ij ≤(nm)(1-D ij );

[0136] Where: m is the number information of the configured feeder terminal devices.

[0137] The virtual power flow lower limit constraint of the virtual branch is constructed by the following eleventh formula:

[0138] 0≤f o,j ≤(nm)(l o,j ).

[0139] In addition, in one embodiment, referring to Figure 4 ,exist Figure 1 After step S14 in the illustrated embodiment, the following steps are also included but not limited to:

[0140] Step S41, obtaining first measurement data and preset basic data information of all measurement devices, and performing a normalization operation on the first measurement data according to the basic data information to obtain a normalization result of the first measurement data;

[0141] Step S42, obtaining a characteristic curve of the first measurement data, determining an optimal solution of the cluster center and the number of clusters in the normalized result according to a preset first function and a first algorithm, and identifying the first measurement data according to the optimal solution;

[0142] Step S43, obtaining the data category, sample number, clustering parameter and fuzzy matrix of the first measurement data, and constructing an indicator function of the first measurement data according to the data category, sample number, clustering parameter, cluster center, cluster number and fuzzy matrix;

[0143] Step S44, obtaining a probability partition matrix of the first measurement data and a normalized sample number of a normalized result, constructing a membership matrix according to the probability partition matrix and the normalized sample number, and confirming the relevance of the first measurement data according to the confirmed data category;

[0144] Step S45, replacing the fuzzy matrix with the membership matrix, and introducing the probability partition matrix into the first function, wherein the fuzzy matrix meets the normalization condition, and obtaining the normalization constraint of the first measurement data;

[0145] Step S46: obtaining a clustering result of the first measurement data according to the normalization constraint and the indicator function.

[0146] It should be noted that, due to the huge scale of measurement data in the low-voltage distribution station area, when the PFCM clustering algorithm is used to cluster the data, the clustering problem of the PFCM algorithm can be effectively solved by setting an indicator function.

[0147] The indicator function is expressed by the following twelfth formula:

[0148]

[0149] Among them: c is the data category; n is the number of samples; r is the clustering parameter; b i is the i-th cluster center; u ij is the fuzzy matrix.

[0150] In the set of data categories: the smaller the value of V(c), the greater the correlation between the data. To obtain the minimum V(c) value, the membership matrix u′ is used ij Replace the original fuzzy matrix and express it through the following thirteenth formula.

[0151]

[0152] Where: t ij is the probability partition matrix; k is the normalized sample number.

[0153] The indicator function contains the fuzzy matrix u′ ij = V(c) and introduce probability partition t ij Matrix. To blur u′ ij The normalization condition is met and expressed by the following fourteenth formula:

[0154]

[0155] Then u′ ij The matrix complies with the normalization constraint and is expressed by the following formula 15.

[0156]

[0157] Furthermore, the indicator function of the PFCM algorithm can achieve the optimal number of clusters through the probability partition matrix while meeting the membership normalization constraint.

[0158] Particle Swarm Optimization:

[0159] Particle selection follows the principle of arbitrariness and seeks the best answer in a continuous iterative process. The iterative process produces two optimal solutions, namely the local optimal solution and the overall optimal solution. Particles are continuously replaced according to these two optimal solutions.

[0160] The search area is d-dimensional, the number of particles in the population is n, and the vector of the i-th particle in the population in d-dimensional is X i =(x i1 ,x i2 ,L,x id ), flight speed V i =(v i1 ,v i2 ,L,v id ), the local optimal solution is the best position found by the particle, which is expressed by the following sixteenth formula:

[0161] P best =(p i1 ,p i2 ,L,p id );

[0162] The global optimal solution is the best position found by the population, which is expressed by the following seventeenth formula:

[0163]

[0164] x id =p id +v id .

[0165] Among them, β is the inertia weight; f1 and f2 are acceleration factors; r1 and r2 are arbitrary numbers in the interval [0,1].

[0166] In addition, in one embodiment, referring to Figure 5 ,exist Figure 1 Step S15 of the illustrated embodiment also includes but is not limited to the following steps:

[0167] Step S51, obtaining the ring adjacency matrix, degree matrix, first output value and activation function of the low-voltage distribution station area, and constructing a GCN model according to the ring adjacency matrix, degree matrix, first output value and activation function;

[0168] Step S52, obtaining the reset gate, update gate, hidden layer and training weight of the GCN model at the first moment, and using the reset gate, update gate, hidden layer and training weight of the GRU model to process the timing information of the low-voltage distribution station area;

[0169] Step S53, confirming whether there is bad data in the second measurement data according to the second algorithm;

[0170] Step S54, when there is no bad data in the second measurement data, directly establish a measurement database;

[0171] Step S55: when bad data exists in the second measurement data, reconstruct the second measurement data according to the GCN model and the GRU model.

[0172] It should be noted that if Fig. 9 As shown in the figure, first, the distribution network measurement information with time series characteristics is taken as input, and the topological structure of the distribution network is analyzed through a convolutional network to obtain the spatial characteristics of multiple low-voltage distribution stations; secondly, the measurement information with spatial characteristics is input into the gated recurrent unit to obtain additional time characteristics; finally, the measurement information is reconstructed through the fully connected layer output.

[0173] The specific process of spatial correlation modeling of low-voltage distribution station areas is as follows:

[0174] Topological structure, the circuit topology is recorded as G = {V, E}, where V = {v1, ···vN} is a set of nodes with a number of N; E = {e1, L, e M} is a set of edges with the number M. Let its adjacency matrix be A∈{0,1} N×N , if A ij =1, indicating that there is a branch from vi to vj, otherwise it means that there is no such branch.

[0175] Feature Matrix The measurement information of nodes and branches is taken as node features and recorded as Where P represents the number of node features, Represents the characteristics at time k.

[0176] Reference Fig.10 , given the adjacency matrix and feature matrix, the filter is constructed in the Fourier domain through the convolutional network model. The filter captures the spatial features between nodes through its first-order neighborhood, and then the convolutional network model is constructed by stacking multiple convolutional layers for nonlinear transformation, which is expressed by the following eighteenth formula:

[0177]

[0178] in, represents a cyclic adjacency matrix; represents the degree matrix; H (l) is the first output value; θ (l) is the weight matrix of layer l; σ represents the sigmoid activation function. It not only has the information of its own node, but also integrates the characteristics of adjacent nodes. In this embodiment, a two-layer GCN model is adopted, which is expressed by the following nineteenth formula:

[0179]

[0180] Among them, W0∈R P×H Represents the weight matrix from the input layer to the hidden layer, P is the length of the feature matrix, H is the number of units in the hidden layer; W1∈R H×T represents the weight matrix from the hidden layer to the output layer; f(X G ,A)∈R N×T Represents the state output with a predicted length of T; Relu represents the activation function.

[0181] Modeling time dependencies:

[0182] Reference Fig.11 , the gated cycle unit is used to process the timing information of the low-voltage distribution station area, which is expressed by the following twentieth formula:

[0183]

[0184] Among them, r k 、u k 、c k They are the reset gate, update gate, and hidden layer at time k respectively; W is the training weight during the training process.

[0185] In order to simultaneously obtain the temporal and spatial correlation of electrical data in low-voltage distribution substations, a T-GCN model based on a convolutional network model and a gated recurrent unit is proposed. Figure X As shown, the left side is the function of spatiotemporal measurement data reconstruction, and the right side is the T-GCN unit structure. hk represents the output at time k, and GCN is a graph convolution process, which is expressed by the following formula:

[0186]

[0187] Where: b are the deviations during the training process.

[0188] In addition, in one embodiment, referring to Figure 6 ,exist Figure 1 Step S15 of the illustrated embodiment also includes but is not limited to the following steps:

[0189] Step S61, obtaining the first measurement sequence number, the first sampling period and the first measurement value of the branch monitoring unit in the low-voltage distribution station area after the second moment;

[0190] Step S62, performing measurement data fusion on the branch monitoring unit according to the first measurement sequence number, the first sampling period and the first measurement value;

[0191] Step S63, obtaining the second measurement sequence number, the second sampling period and the second measurement value of the photovoltaic switch collection device in the low-voltage distribution station area after the second moment in the low-voltage distribution station area;

[0192] Step S64, performing measurement data fusion on the photovoltaic switch acquisition device according to the second measurement sequence number, the second sampling period and the second measurement value;

[0193] Step S65, obtaining the third measurement sequence number, the third sampling period and the third measurement value of the multi-function meter in the low-voltage distribution station area after the second moment;

[0194] Step S66, fusing the measurement data of the multi-function meter according to the third measurement sequence number, the third sampling period and the third measurement value;

[0195] Step S67, obtaining a fourth measurement sequence number, a fourth sampling period, and a fourth measurement value of the smart meter in the low-voltage distribution station area after the second moment in the low-voltage distribution station area;

[0196] Step S68, performing measurement data fusion on the smart meter according to the fourth measurement sequence number, the fourth sampling period and the fourth measurement value.

[0197] It should be noted that, in this embodiment, the branch monitoring unit measurement data is integrated based on average recursive filtering, which is expressed by the following formula 22:

[0198]

[0199] Wherein, K1 is the first measurement sequence number of the branch monitoring unit after time t; T is the first sampling period measured by the branch monitoring unit; is the first measurement value of the branch monitoring unit at time t+kT.

[0200] Based on average recursive filtering, the measurement data of the photovoltaic switch terminal acquisition equipment is fused and expressed by the following formula 23;

[0201]

[0202] Wherein, K2 is the second measurement sequence number of the photovoltaic switch terminal acquisition device after time t; T is the second sampling period measured by the photovoltaic switch terminal acquisition device; is the third measurement value at time t+kT.

[0203] The multi-function meter measurement data is fused based on average recursive filtering, which is expressed by the following formula 24:

[0204]

[0205] Where: K3 is the third measurement sequence number of the multi-function meter after time t; T is the third sampling period measured by the multi-function meter; is the third measurement value at time t+kT.

[0206] The smart meter measurement data is fused based on average recursive filtering, which is expressed by the following formula 25:

[0207]

[0208] Where: K4 is the fourth measurement sequence number of the smart meter after time t; T is the fourth sampling period of the smart meter measurement; is the fourth measurement value at time t+kT.

[0209] In addition, in one embodiment, referring to Figure 7 ,exist Figure 1 Step S13 of the illustrated embodiment also includes but is not limited to the following steps:

[0210] Step S71, obtaining a horizontal component, a vertical component, a first smoothing parameter, a second smoothing parameter, and a first predicted value and a first estimated value of a state quantity at a third moment in a low-voltage distribution station area;

[0211] Step S72, constructing a state equation of the low voltage distribution station area according to the horizontal component, the vertical component, the first smoothing parameter, the second smoothing parameter, the first predicted value and the second predicted value;

[0212] Step S73, obtaining the phase angle difference, active injection power, reactive injection power, mutual admittance between multiple nodes, active power at the beginning of the line, and reactive power at the beginning of the line of multiple nodes in the low-voltage distribution station area;

[0213] Step S74, constructing a measurement equation for the low-voltage distribution station area according to the phase angle difference, active injection power, reactive injection power, mutual admittance, active power at the beginning of the line, and reactive power at the beginning of the line;

[0214] Step S75, obtaining the sampling point mean, sampling point weight, proportional correction factor, free parameter, influencing factor, state quantity mean and first covariance of the low voltage distribution station area;

[0215] Step S76, selecting Sigma points according to the sampling point mean, sampling point weight, proportional correction factor, free parameter, influencing factor, state quantity mean and first covariance, and constructing a system prediction step model according to the Sigma points;

[0216] Step S77, obtaining the state quantity mean, state quantity covariance, Sigma point measurement value, third prediction value, measurement prediction covariance and second covariance according to the system prediction unit model, wherein the second covariance is the difference between the first prior state estimation value and the measurement prediction covariance of the low-voltage distribution station area;

[0217] Step S78, constructing a measurement update step model according to the state quantity mean, the state quantity covariance, the Sigma point measurement value, the third prediction value, the measurement prediction covariance and the second covariance;

[0218] Step S79, obtaining a second a priori prediction value, a third a priori prediction value, an actual measurement value and a posteriori estimation value of the low voltage distribution station area;

[0219] Step S80, constructing a state deduction model according to the state equation, the measurement equation, the prediction step model, the update step model, the second a priori prediction value, the third a priori prediction value, the actual measurement value and the posterior estimation value.

[0220] It should be noted that the state equation for dynamic state estimation of the low-voltage distribution station area is established by the following formula 26:

[0221]

[0222] in: is the first predicted value of the state quantity at the third moment; is the first estimated value of the state quantity at the third moment; s k is the horizontal component; b k is the vertical component; α H is the first smoothing parameter, β His the second smoothing parameter. The state quantity when estimating the state of the low-voltage distribution substation is the voltage amplitude of each node. The state quantity of node i can be expressed as x i =[V i ], the state quantity x of the system can be expressed as x=[x1,…,x N ]. Among them, V i is the voltage amplitude of node i; N is the number of nodes in the low-voltage distribution station area.

[0223] The measurement equation for dynamic state estimation of low-voltage distribution station area is established and expressed by the following formula 27:

[0224]

[0225] Where: θ ij is the phase angle difference between the two nodes, θ ij =θ i -θ j ;P i is the active injection power of node i; Q i is the reactive injection power of node i; G ij +jB ij is the mutual admittance between node i and node j; P ij is the active power at the beginning of line ij, and the direction is defined as flowing from i to j is positive, and flowing from j to i is negative; Q ij is the reactive power at the starting end of line ij, and its direction is the same as above.

[0226] The Sigma point is selected based on the proportional correction symmetric sampling method, which is expressed by the following formula 28 and formula 29:

[0227]

[0228] Where: m is the sampling point mean of the state quantity; P is the covariance of the state quantity; W m and W c are the weights of each sampling point for calculating the mean and covariance during untraceable transformation; α is the proportional correction factor; κ is a free parameter used to capture the high-order moment information of a given distribution; β is an influencing factor, which contains the high-order moment information of the prior distribution.

[0229] The model of the system prediction step is established by the following formula 30, formula 31 and formula 32:

[0230]

[0231]

[0232] in: is a set of Sigma points obtained according to the sampling scheme and the mean and covariance of the state quantity at the current moment, is the sampling point for the next moment.

[0233] The model establishment of the measurement update step is expressed by the following Formula 33, Formula 34, Formula 35, Formula 36, ​​Formula 37, Formula 38 and Formula 39:

[0234]

[0235] in, is the mean value of the state quantity obtained in the prediction step, is the covariance of the state quantity obtained in the prediction step, is the measured value of each Sigma point, The third predicted value of the k-time measurement value is obtained by weighted summation of each measurement value, S k is the measurement prediction covariance, C k is the second covariance between the prior state estimate and the system measurement prediction value.

[0236] The dynamic state estimation model of low-voltage distribution substation based on UKF is expressed by the following formula:

[0237]

[0238] in: is the second a priori prediction value of the state quantity of the low-voltage distribution station area at the third moment, that is, the a priori prediction value of the node voltage amplitude and phase angle; is the third a priori prediction value of the low-voltage distribution station area quantity measurement at the third moment, that is, the a priori prediction value of the node voltage amplitude and the node and branch power; k The actual measurement value of the measurement uploaded by the measurement device at the third moment; It is the a posteriori estimate of the state quantity of the low-voltage distribution station area at the third moment.

[0239] like Figure 8 As shown, Figure 8 The present invention also provides a data state deduction device for a distribution station area, including:

[0240] The processor 801 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0241] The memory 802 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 802, and the processor 801 calls and executes the distribution station area terminal configuration and dynamic deduction method of the embodiment of this application;

[0242] Input / output interface 803, used to implement information input and output;

[0243] The communication interface 804 is used to realize the communication interaction between the present apparatus and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0244] A bus 805 that transmits information between the various components of the device (e.g., the processor 801, the memory 802, the input / output interface 803, and the communication interface 804);

[0245] The processor 801 , the memory 802 , the input / output interface 803 and the communication interface 804 are connected to each other in communication within the device via a bus 805 .

[0246] An embodiment of the present application also provides an electronic device, comprising the distribution station area data status deduction device as described above.

[0247] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned distribution station area terminal configuration and dynamic deduction method is implemented.

[0248] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are implemented to be located in one place, or may also be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0249] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0250] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions under the shared conditions without violating the spirit of the present invention. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A method for terminal configuration and dynamic deduction in a distribution area, characterized in that: include: Acquire a first topological structure of a low-voltage distribution station area, an installation location of a measuring device, electrical information collected by the measuring device, and a terminal configuration node, wherein the low-voltage distribution station area is provided with a feeder terminal device; According to the first topology structure, the installation location and the electrical information supplement zero injection node, the low voltage monitoring unit is initialized, and the virtual load of the low voltage distribution station area is assigned according to the terminal configuration node to obtain a measurement diagram of the low voltage distribution station area; Constructing an observable tree search method for the low-voltage distribution station area to perform observability analysis on the measurement diagram to determine the observable range of the low-voltage distribution station area, and obtaining a configuration scheme of the feeder terminal device according to the observable range; Acquire first measurement data of the low-voltage distribution station area according to the configuration scheme, perform time-scale alignment on the first measurement data, and obtain second measurement data; A measurement database of the low-voltage distribution station area is established according to the second measurement data, and a state deduction model is established according to the measurement database to complete a dynamic real-time evaluation of the low-voltage distribution station area.

2. The method for terminal configuration and dynamic deduction in a distribution area according to claim 1, characterized in that: The observable tree search method for constructing the low voltage distribution station area performs observability analysis on the measurement diagram, including: Confirm multiple equivalent connection nodes in the measurement graph, assign edge weights to the equivalent connection nodes, obtain a first node set, a first edge weight value, and the number of nodes of the measurement graph, and when the first node set and the first edge weight value are equal to the number of nodes, obtain the maximum observable range of the measurement graph; When the node set and the edge weight value are not equal to the number of nodes, confirm multiple first isolated points in the measurement diagram, search for a first boundary connected to the first isolated point, and when there is reactive power flow measurement at the first boundary, obtain the second edge weight value and the second node set until all the first isolated points are searched, and when the second variable weight value and the second node set are equal to the number of nodes, obtain the maximum observable range of the measurement diagram; Confirming a plurality of second isolated points in the measurement diagram, and when the reactive power flow measurement does not exist between the second isolated point and the first boundary, acquiring a first reactive injection measurement node and a third node set in the nodes; According to the first reactive injection node and the third node set, a fourth node set of the nodes is obtained; when there is no reactive power flow measurement at all the boundaries, a second reactive injection node is obtained; when there is a second boundary connected to the second reactive injection node, and the reactive power flow measurement exists between the second boundary and the second reactive injection node, a fourth node set is obtained; when there is no reactive power flow measurement between the second boundary and the second reactive injection node, the search is continued until all the nodes are searched; When the first node set, the second node set, the third node set and the fourth node set are all equal to the number of nodes, the maximum observable range of the measurement graph is obtained; When the low-voltage distribution station is divided into a plurality of observable areas, the plurality of observable areas output their respective maximum observable ranges.

3. The method for terminal configuration and dynamic deduction in distribution area according to claim 1, characterized in that: The step of obtaining a configuration scheme of the feeder terminal device according to the observable range includes: Acquire a virtual power flow value of the feeder terminal device, and obtain an objective function of the feeder terminal device according to the virtual power flow value; Acquire the virtual power flow of the low-voltage distribution substation area, and construct the virtual power flow constraint of the low-voltage distribution substation area according to the objective function and the virtual power flow; Acquire the branch set of the low-voltage distribution station area and the configuration status information of the feeder terminal device, and construct a configuration quantity constraint of the feeder terminal device according to the configuration status information of the branch set; Acquire the virtual load of the node of the low-voltage distribution substation, calculate the line virtual power flow of the low-voltage distribution substation according to the virtual load and the virtual power flow, and construct the connectivity constraint of the low-voltage distribution substation according to the virtual load, the virtual power flow and the line virtual power flow; Acquire virtual branch variable information of the low-voltage distribution station area and quantity information of the feeder terminals, and construct radial constraints of the low-voltage distribution station area according to the virtual branch variable information and the quantity information; The upper limit constraint of the branch virtual power flow in the low-voltage distribution station area is constructed according to the quantity information, the configuration quantity constraint and the connectivity constraint, and the lower limit constraint of the branch virtual power flow in the low-voltage distribution station area is constructed according to the virtual power flow constraint, the quantity information and the radial constraint.

4. The method for terminal configuration and dynamic deduction in a distribution area according to claim 1, characterized in that: After obtaining the first measurement data of the low-voltage distribution station area according to the configuration scheme, the method further includes: Acquire the first measurement data and preset basic data information of all the measurement devices, and perform a normalization operation on the first measurement data according to the basic data information to obtain a normalization result of the first measurement data; Acquire a characteristic curve of the first measurement data, wherein the characteristic curve determines an optimal solution of cluster centers and the number of clusters in the normalization result according to a preset first function and a first algorithm, and identify the first measurement data according to the optimal solution; Acquire the data category, sample number, clustering parameter and fuzzy matrix of the first measurement data, and construct an indicator function of the first measurement data according to the data category, sample number, clustering parameter, cluster center, cluster number and fuzzy matrix; Obtaining a probability partition matrix of the first measurement data and a normalized sample number of the normalized result, constructing a membership matrix according to the probability partition matrix and the normalized sample number, and confirming the relevance of the first measurement data according to confirming the data category; The membership matrix replaces the fuzzy matrix, and introduces the probability partition matrix into the first function, wherein the fuzzy matrix meets the normalization condition, so as to obtain the normalization constraint of the first measurement data; A clustering result of the first measurement data is obtained according to the normalization constraint and the indicator function.

5. The method for terminal configuration and dynamic deduction in a distribution area according to claim 1, characterized in that: The step of establishing a measurement database of the low-voltage distribution station area according to the second measurement data comprises: Obtaining a ring adjacency matrix, a degree matrix, a first output value, and an activation function of the low-voltage distribution station area, and constructing a GCN model according to the ring adjacency matrix, the degree matrix, the first output value, and the activation function; Obtaining the reset gate, update gate, hidden layer and training weight of the GCN model at the first moment, and using the reset gate, update gate, hidden layer and training weight of the GRU model to process the timing information of the low-voltage distribution station area; confirming whether there is bad data in the second measurement data according to a second algorithm; When the bad data does not exist in the second measurement data, directly establishing a measurement database; When the bad data exists in the second measurement data, data reconstruction is performed on the second measurement data according to the GCN model and the GRU model.

6. The method for terminal configuration and dynamic deduction in a distribution area according to claim 1, characterized in that: The performing time-scale alignment on the first measurement data includes: Obtaining the first measurement sequence number, the first sampling period and the first measurement value of the branch monitoring unit of the low-voltage distribution station area after the second moment of the low-voltage distribution station area; Performing measurement data fusion on the branch monitoring unit according to the first measurement sequence number, the first sampling period and the first measurement value; Obtaining the second measurement sequence number, the second sampling period and the second measurement value of the photovoltaic switch collection device in the low-voltage distribution station area after the second moment in the low-voltage distribution station area; Performing measurement data fusion on the photovoltaic switch acquisition device according to the second measurement sequence number, the second sampling period and the second measurement value; Obtaining a third measurement sequence number, a third sampling period, and a third measurement value of the multi-function meter in the low-voltage distribution station area after the second moment in the low-voltage distribution station area; Performing measurement data fusion on the multi-function meter according to the third measurement sequence number, the third sampling period and the third measurement value; Obtaining a fourth measurement sequence number, a fourth sampling period, and a fourth measurement value of the smart electric meter in the low-voltage distribution station area after the second moment in the low-voltage distribution station area; The measurement data of the smart meter is fused according to the fourth measurement sequence number, the fourth sampling period and the fourth measurement value.

7. The method for configuring and dynamically deducing distribution area terminals according to claim 1, characterized in that: The establishing of a state deduction model according to the measurement database comprises: Obtaining a horizontal component, a vertical component, a first smoothing parameter, a second smoothing parameter, and a first predicted value and a first estimated value of a state quantity at a third moment of the low-voltage distribution station area; Constructing a state equation of the low voltage distribution station area according to the horizontal component, the vertical component, the first smoothing parameter, the second smoothing parameter, the first predicted value and the second predicted value; Obtaining the phase angle difference, active injection power, reactive injection power, mutual admittance between the multiple nodes, active power at the beginning of the line, and reactive power at the beginning of the line of multiple nodes in the low-voltage distribution station area; Constructing a measurement equation of the low-voltage distribution station area according to the phase angle difference, the active injection power, the reactive injection power, the mutual admittance, the active power at the beginning of the line and the reactive power at the beginning of the line; Obtaining the sampling point mean, sampling point weight, proportional correction factor, free parameter, influencing factor, state quantity mean and first covariance of the low voltage distribution station area; Selecting a Sigma point according to the sampling point mean, the sampling point weight, the proportional correction factor, the free parameter, the influencing factor, the state quantity mean and the first covariance, and constructing a system prediction step model according to the Sigma point; According to the system prediction unit model, the state quantity mean, state quantity covariance, Sigma point measurement value, third prediction value, measurement prediction covariance and second covariance are obtained, wherein the second covariance is the difference between the first prior state estimation value of the low-voltage distribution station area and the measurement prediction covariance; Constructing a measurement update step model according to the state quantity mean, the state quantity covariance, the Sigma point measurement value, the third predicted value, the measurement prediction covariance and the second covariance; Obtaining a second a priori prediction value, a third a priori prediction value, an actual measurement value and a posteriori estimation value of the low voltage distribution station area; A state deduction model is constructed according to the state equation, the measurement equation, the prediction step model, the update unit model, the second priori prediction value, the third priori prediction value, the actual measurement value and the posterior estimation value.

8. A data status deduction device for a distribution station area, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the distribution station area terminal configuration and dynamic deduction method as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes the distribution station area data status deduction device as described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the distribution station area terminal configuration and dynamic deduction method as described in any one of claims 1 to 7.

Images